Cold-sensing regulates Drosophila growth through insulin-producing cells

Abstract

Across phyla, body size is linked to climate. For example, rearing fruit flies at lower temperatures results in bigger body sizes than those observed at higher temperatures. The underlying molecular basis of this effect is poorly understood. Here we provide evidence that the temperature-dependent regulation of Drosophila body size depends on a group of cold-sensing neurons and insulin-producing cells (IPCs). Electrically silencing IPCs completely abolishes the body size increase induced by cold temperature. IPCs are directly innervated by cold-sensing neurons. Stimulation of these cold-sensing neurons activates IPCs, promotes synthesis and secretion of Drosophila insulin-like peptides and induces a larger body size, mimicking the effects of rearing the flies in cold temperature. Taken together, these findings reveal a neuronal circuit that mediates the effects of low temperature on fly growth.

Figure 1. IPCs are required for the effect of low temperature on pupal size.

(a) Pupariation time of w¹¹¹⁸ at 18 °C was later than at 25 °C (n=5). (b) w¹¹¹⁸ pupal size was greater at 18 °C than at 25 °C (n=28 for females; n=22 for males). (c) Absorbance of iodo–starch reactions at 580 nm using residue food from w¹¹¹⁸ fly cultures at 25 or 18 °C. Food starch remaining after 25 °C culture was less than after 18 °C culture (n=3). See Methods section for further details. (d) Pupariation time of larvae expressing NaChBac with dilp2-Gal4 was later than that of controls both at 25 °C (n=7) and at 18 °C (n=8). (e,f) Pupal size of flies expressing NaChBac with dilp2-Gal4 was larger than that of controls at 18 °C (n=24 for both females and males) as shown in e, but not at 25 °C (n=24 for females; n=33 for males) as shown in f. (g) Absorbance of iodo–starch reaction at 580 nm using residue food after culturing flies expressing NaChBac with dilp2-Gal4. Food starch remaining for flies with IPCs hyperactivated was not different from that of controls (n=3). (h) Pupariation time of larvae expressing Kir2.1 by dilp2-Gal4 and control similarly increased at 18 °C as compared with at 25 °C (n=9). (i) Pupal sizes of flies with blocked IPCs cultured at 18 °C were not different from those cultured at 25 °C (P>0.05, n=13 for both females and males); pupal sizes of controls were significantly larger at 18 °C than at 25 °C (n=13). F, females; M, males; AEH, after egg hatching; error bars are s.e.m.; ***P<0.001, Student's t-test.

Figure 2. IPCs are responsive to cold temperature.

(a,b) Ca²⁺ imaging of IPCs responses to a temperature decrease. The responses of two groups of IPCs on each side of the brain are shown as representatives. (c) CaLexA-based imaging of larval IPCs after 24 h culture at 18 °C. (d) Quantification of c (n=7). (e–g) Culturing w¹¹¹⁸ larvae at 18 °C for 6 h increased mRNA expression levels of dilp2, dilp3 and dilp5. Fold changes are relative to conditions at 0 h (n=3). (h) Anti-Dilp2 staining in IPCs of w¹¹¹⁸ larvae raised on poor food was lower after 18 °C treatment. (i) Quantification of h (n=20). (j,k) Dilp2 levels decreased in brain but increased in haemolymph after 6 h of 18 °C treatment. (j) Western blotting of brain and haemolymph Dilp2. (k) Quantification of j (n=3). Scale bars, 50 μm for all; error bars are s.e.m.; **P<0.01, ***P<0.001, Student's t-test or analysis of variance.

Figure 3. 11216-Gal4 labels cold-sensing neurons in larval flies.

(a–c) Expression of 11216-Gal4 and Or83b-RFP in larval head and central nervous system. Arrows indicates the DOG neurons. Note that the two rightmost DOG neurons are overlaid in b. Scale bars, 20 μm. (d,e) Ca²⁺ imaging of 11216-Gal4 neurons in response to a temperature decrease. Imaging of 11216-Gal4 neurons in one representative sample is shown as a heat map (d) and corresponding fluorescence intensity curves (e). Four different groups of neurons are designated a, b, c and d. (f) The scatter plot shows peak responses of 11216-Gal4 neurons to ice water in three representative samples. (g) CaLexA-based imaging of the axonal termini of 11216-Gal4 neurons after 24 h culture at 18 °C. Scale bars, 50 μm. (h) Quantification of g (n=7). (i) Activation of 11216-Gal4 neurons by overexpressing NaChBac-delayed pupariation at both 25 °C (n=6) and 18 °C (n=6). (j) At 25 °C, pupal sizes of flies with 11216-Gal4 neurons activated by overexpressing NaChBac were larger than in parental controls for females but not for males (n=29 for females; n=21 for males). (k) At 18 °C, an increase in pupal size was seen in both females and males (n=21 for females; n=19 for males). Fold changes are relative to the average pupal size of female 11216-Gal4 flies. (l) Absorbance of iodo–starch reaction at 580 nm using residue food after culturing flies expressing UAS-NaChBac by 11216-Gal4. Food starch after culturing 11216-Gal4 neurons-activated flies was not different from that of controls (n=3). (m,n) Optogenetically activating 11216-Gal4 neurons delayed pupariation (m) and enhanced pupal size in both females and males (n; for pupariation, n=4; for pupal size, n=21 for females, n=18 for males.) (o,p) Pupariation time (o) and pupal sizes (p) of larvae with ectopic expression of TNTG by 11216-Gal4 and of controls increased similarly at 18 °C as compared with at 25 °C (for pupariation, n=8; for pupal size, n=17 for both females and males). F, females; M, males; AEH, after egg hatching; error bars are s.e.m.; **P<0.01, ***P<0.001, Student's t-test.

Figure 4. Larval 11216-Gal4 neurons directly interact with IPCs.

(a–d) A strong GRASP signal was seen between 11216-Gal4 neurons and IPCs but no detectable signal was seen in controls. Magenta, anti-CD4 (labels both 11216-Gal4 neurons and IPCs); green, GFP (GRASP signal). Scale bars, 10 μm. (e) Ca²⁺imaging of IPCs when 11216-Gal4 neurons expressing Chrimson were activated by 620 nm red light. The IPCs on both sides of the brain showed asynchronous responses. Groups 1 and 2 are two groups of IPCs on each side of the brain. (f) The histogram shows the average maximum increase in fluorescence intensity in experimental and control samples (n=7). (g) Ca²⁺ imaging of the response of larval IPCs to ice water in flies with 11216-Gal4 neurons ablated with Diphtheria toxin (DTI). In a representative sample, two groups of IPCs on each side of the brain showed strong responses. (h–m) Larvae with 11216-Gal4 neurons activated by NaChBac showed higher dilp2 and dilp5 mRNA levels at 18 °C but not 25 °C, while dilp3 mRNA was always higher than controls at both 18 and 25 °C. Fold changes are relative to mRNA levels in 11216-Gal4 larvae (n=3). (n–o) Anti-DILP2 staining signal in poor food-raised flies with 11216-Gal4 neurons activated by NaChBac was lower than in controls. Fold changes are relative to level in 11216-Gal4 (n=17). Scale bars, 50 μm. (p,q) Dilp2 levels decreased in brain but increased in haemolymph when 11216-Gal4 neurons were activated by NaChBac. (p) Western blotting of brain and haemolymph Dilp2. (q) Quantification of p (n=3). Error bars are s.e.m.; *P<0.05, **P<0.01, ***P<0.001, Student's t-test.